1.2 “Meiji”

Autoprefixer uses Can I Use data. Previous versions contain Can I Use dump inside npm package and I had special script to update this data from Can I Use repo. I manually run this scripts every day. It was a hack, not a good solution.

Few month ago @Fyrd started to publish official caniuse-db npm package. Autoprefixer 1.2 start to use this official data instead of local copy.

This changes made Autoprefixer’s maintaining much easier, but also it gives a few benefits to end-users:

1. Autoprefixer stops to use strange 1.1.20140605 versions with dump date in minor part. Now it will be normal versions like 1.2.0.
2. You can update Can I Use data separetly from Autoprefixer’s code. It is important for big companies, which have long test period on every code changes. Now you can have up-to-date browsers data and don’t be worry to broke styles.
3. Autoprefixer extremely cares about actual prefix data. In previous version there was day delay between Can I Use updates and Autoprefixer release. Now you will get faster updates directly from Can I Use author. Don’t forget to call npm update caniuse-db.